Re: [Phys-l] The "why" questions

[Hugh Haskell <haskellh@verizon.net>]

At 05:12 -0700 11/29/2010, John Denker wrote:
> The "why" in those situations can be replaced by "how do we know".
>
>  a) We know that if there is an unbalanced force there must be an
>   acceleration.  Given the force we can calculate the acceleration
>
>  b) We also know that if there is an acceleration, there must be an
>   unbalanced force.  Given the acceleration we can calculate the force.
>
> Let's be clear:  There can be a cause for the knowing, even though
> there is not a cause for the acceleration.  In some situations (e.g.
> the diver) know the force in advance and calculate the acceleration. 
> In other situations (e.g. a centrifuge) you know the acceleration in
> advance and calculate the force.
>
> > >  But I know I do say things like:  "An unbalanced force causes an
> > >  object to accelerate."
>
> I used to say things like that, but I've pretty much trained myself
> to not do so anymore.  Instead I might say an unbalanced force is
> /associated with/ an acceleration.  The point is that association
> is symmetric, unlike causation which is asymmetric.  F=ma is an
> equality, and equality is reflective, symmetric, and transitive.
>
> I might also say that given an unbalanced force, we know the object
> will accelerate.  I might even say that because there is an unbalanced
> force, we know the object will accelerate.  (There is a cause for the
> knowing, even though there is not a cause for the acceleration.)
>
> Sometimes I say "because" when I shouldn't, but I treat this as a
> mistake just like any other mistake, and I edit it out if I can.

I think we are making too much of a distinction between "how" and 
"why." All that we ever do is descriptive, and no hows or whys are 
ever assumed if we stick to our actual activities. What we do is 
build models of what we observe and then test them to see if they 
make correct predictions. All we ever have to work with is a black 
box with some inputs and outputs, and we have to guess what is in the 
black box. If we guess well, then the outputs of the black box will 
match what we have predicted based on our assumed interior structure, 
more or less. If we don't then they won't.

Our use of hows and whys is merely a shorthand to enable us to carry 
on a conversation. And if we use the word how or why to explain the 
next order of the approximation to reality that our black box 
inevitably is, it only means that "our prediction is that if we put A 
into the box, we get B from the output port. If one is to ask why 
this is, then we must go to a higher order of the approximation--a 
more complex model of what we think might be inside the black box. At 
some point (and this is often a moving target), we have no further to 
go in the complexity of our model and we must then say, "we don't 
know--all we can say at this point is this is the way is is, or more 
precisely, this is the way it appears to be within the limits of our 
model." None of this implies and cause & effect, or even any reality 
of our model. All it says is that if we do A, then we get B back from 
our model. Nothing more.

This concept is much more completely explained in a book I have 
recommended to this list before: "Doing Physics: How Physicists Take 
Hold of the World," by Martin Krieger (Indiana University Press, 
1992). Physicists don't actually figure out how the world works, they 
only build models that attempt to simulate how the world *might* 
work. In this context it is absolutely clear that "becauses," "whys," 
"hows" and all those similar words which imply cause-and-effect are 
mere place holders for the much more complex descriptions of the 
human-constructed models. I think that the sooner we make this clear 
to students the less confused they are likely to be by what we do. I 
think that the upshot is that physicists are all Platonists in the 
final analysis.

Hugh
--

Hugh Haskell
mailto:hugh@ieer.org
mailto:haskellh@verizon.net

It isn't easy being green.

		--Kermit Lagrenouille